Interpreting Lion Behaviour with Nonparametric Probabilistic Programs

Neil Dhir, Matthijs Vákár, Matthew Wijers, Andrew Markham, Frank Wood, Paul Trethowan, Byron Du Preez, Andrew Loveridge, David MacDonald
Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence, PMLR R15:451-460, 2017.

Abstract

We consider the problem of unsupervised learning of meaningful behavioural segments of high-dimensional time-series observations, collected from a pride of African lions1. We demonstrate, by way of a probabilistic programming system (PPS), a methodology which allows for quick iteration over mod- els and Bayesian inferences, which enables us to learn meaningful behavioural segments. We introduce a new Bayesian nonparametric (BNP) state-space model, which extends the hierarchical Dirichlet process (HDP) hidden Markov model (HMM) with an explicit BNP treatment of duration distributions, to deal with different levels of granularity of the latent be- havioural space of the lions. The ease with which this is done exemplifies the flexibility that a PPS gives a scientist2. Furthermore, we combine this approach with unsupervised fea- ture learning, using variational autoencoders.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR15-dhir17a, title = {Interpreting Lion Behaviour with Nonparametric Probabilistic Programs}, author = {Dhir, Neil and V{\'a}k{\'a}r, Matthijs and Wijers, Matthew and Markham, Andrew and Wood, Frank and Trethowan, Paul and Preez, Byron Du and Loveridge, Andrew and MacDonald, David}, booktitle = {Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence}, pages = {451--460}, year = {2017}, editor = {Elidan, Gal and Kersting, Kristian}, volume = {R15}, series = {Proceedings of Machine Learning Research}, month = {11--15 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r15/main/assets/dhir17a/dhir17a.pdf}, url = {https://proceedings.mlr.press/r15/dhir17a.html}, abstract = {We consider the problem of unsupervised learning of meaningful behavioural segments of high-dimensional time-series observations, collected from a pride of African lions1. We demonstrate, by way of a probabilistic programming system (PPS), a methodology which allows for quick iteration over mod- els and Bayesian inferences, which enables us to learn meaningful behavioural segments. We introduce a new Bayesian nonparametric (BNP) state-space model, which extends the hierarchical Dirichlet process (HDP) hidden Markov model (HMM) with an explicit BNP treatment of duration distributions, to deal with different levels of granularity of the latent be- havioural space of the lions. The ease with which this is done exemplifies the flexibility that a PPS gives a scientist2. Furthermore, we combine this approach with unsupervised fea- ture learning, using variational autoencoders.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Interpreting Lion Behaviour with Nonparametric Probabilistic Programs %A Neil Dhir %A Matthijs Vákár %A Matthew Wijers %A Andrew Markham %A Frank Wood %A Paul Trethowan %A Byron Du Preez %A Andrew Loveridge %A David MacDonald %B Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2017 %E Gal Elidan %E Kristian Kersting %F pmlr-vR15-dhir17a %I PMLR %P 451--460 %U https://proceedings.mlr.press/r15/dhir17a.html %V R15 %X We consider the problem of unsupervised learning of meaningful behavioural segments of high-dimensional time-series observations, collected from a pride of African lions1. We demonstrate, by way of a probabilistic programming system (PPS), a methodology which allows for quick iteration over mod- els and Bayesian inferences, which enables us to learn meaningful behavioural segments. We introduce a new Bayesian nonparametric (BNP) state-space model, which extends the hierarchical Dirichlet process (HDP) hidden Markov model (HMM) with an explicit BNP treatment of duration distributions, to deal with different levels of granularity of the latent be- havioural space of the lions. The ease with which this is done exemplifies the flexibility that a PPS gives a scientist2. Furthermore, we combine this approach with unsupervised fea- ture learning, using variational autoencoders. %Z Reissued by PMLR on 04 October 2026.
APA
Dhir, N., Vákár, M., Wijers, M., Markham, A., Wood, F., Trethowan, P., Preez, B.D., Loveridge, A. & MacDonald, D.. (2017). Interpreting Lion Behaviour with Nonparametric Probabilistic Programs. Proceedings of the 33rd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R15:451-460 Available from https://proceedings.mlr.press/r15/dhir17a.html. Reissued by PMLR on 04 October 2026.

Related Material